As the use of the Internet has grown, so has the interest in Internet appliances. Such appliances typically comprise relatively low-cost devices specifically designed for use on-line without the need for a personal computer (PC). Due to their focused application, and to keep costs low, such devices normally have less memory and computing capacity as compared with conventional PCs.
Recently, demand for simplified scanning devices that can be used as Internet appliances has risen. Preferably, such devices would operate as stand alone devices that could be used to scan and transmit images over the Internet without the need for a PC. Such scanning devices could be used, for instance, to e-mail pictures and other images so that the images can be displayed on the recipient's monitor or television.
In that conventional scanners are used as peripheral equipment, the scanner normally receives instructions from a PC under the control of the user. Since conventional scanners typically scan each image at a preset optical resolution, a user may need to manually select a different scanning resolution if the user wishes to change the scale of the image. In situations where the scanner is to be used as an Internet appliance, however, the scanner must have the capability to process the scanned data internally. It is desirable to automate this processing, including scaling, to simplify the use of the scanner. Automatic scaling presents a challenge because the correct scanning resolution of the image, and therefore the correct scaling of the image, is not known until scanning of the entire image is completed. However, full scanning and possible rescanning of the entire image is undesirable, particularly in the Internet appliance context, in that such scanning delays image transmission and requires relatively large amounts of memory and computing power.
From the foregoing, it can be appreciated that it would be desirable to have a scanner which automatically determines the correct scanning resolution for each original image such that the image can be automatically scaled to obtain a full image having high resolution without the need to fully scan and rescan the image.